Paper
14 March 2013 Learning a piecewise linear transform coding scheme for images
Aäron van den Oord, Sander Dieleman, Benjamin Schrauwen
Author Affiliations +
Proceedings Volume 8768, International Conference on Graphic and Image Processing (ICGIP 2012); 876844 (2013) https://doi.org/10.1117/12.2011134
Event: 2012 International Conference on Graphic and Image Processing, 2012, Singapore, Singapore
Abstract
Gaussian mixture models are among the most widely accepted methods for clustering and probability density estimation. Recently it has been shown that these statistical methods are perfectly suited for learning patch-based image priors for various image restoration problems. In this paper we investigate the use of GMM's for image compression. A piecewise linear transform coding scheme based on Vector Quantization is proposed. In this scheme two different learning algorithms for GMM's are considered and compared. Experimental results demonstrate that the proposed techniques outperform JPEG, with results comparable to JPEG2000 for a broad class of images.
© (2013) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Aäron van den Oord, Sander Dieleman, and Benjamin Schrauwen "Learning a piecewise linear transform coding scheme for images", Proc. SPIE 8768, International Conference on Graphic and Image Processing (ICGIP 2012), 876844 (14 March 2013); https://doi.org/10.1117/12.2011134
Lens.org Logo
CITATIONS
Cited by 1 scholarly publication.
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Expectation maximization algorithms

Image compression

JPEG2000

Quantization

Data modeling

Image quality

Image restoration

RELATED CONTENT


Back to Top